Utilize este identificador para referenciar este registo:
https://hdl.handle.net/1822/86333
Título: | On the impact of interpretability methods in active image augmentation method |
Autor(es): | Oliveira Santos, Flavio Arthur Zanchettin, Cleber Matos, Leonardo Nogueira Novais, Paulo |
Palavras-chave: | Data augmentation robustness interpretability |
Data: | 2022 |
Editora: | Oxford University Press |
Revista: | Logic Journal of the IGPL |
Citação: | Flávio Arthur Oliveira Santos, Cleber Zanchettin, Leonardo Nogueira Matos, Paulo Novais, On the Impact of Interpretability Methods in Active Image Augmentation Method, Logic Journal of the IGPL, Volume 30, Issue 4, August 2022, Pages 611–621, https://doi.org/10.1093/jigpal/jzab006 |
Resumo(s): | Robustness is a significant constraint in machine learning models. The performance of the algorithms must not deteriorate when training and testing with slightly different data. Deep neural network models achieve awe-inspiring results in a wide range of applications of computer vision. Still, in the presence of noise or region occlusion, some models exhibit inaccurate performance even with data handled in training. Besides, some experiments suggest deep learning models sometimes use incorrect parts of the input information to perform inference. Active image augmentation (ADA) is an augmentation method that uses interpretability methods to augment the training data and improve its robustness to face the described problems. Although ADA presented interesting results, its original version only used the vanilla backpropagation interpretability to train the U-Net model. In this work, we propose an extensive experimental analysis of the interpretability method's impact on ADA. We use five interpretability methods: vanilla backpropagation, guided backpropagation, gradient-weighted class activation mapping (GradCam), guided GradCam and InputXGradient. The results show that all methods achieve similar performance at the ending of training, but when combining ADA with GradCam, the U-Net model presented an impressive fast convergence. |
Tipo: | Artigo |
URI: | https://hdl.handle.net/1822/86333 |
DOI: | 10.1093/jigpal/jzab006 |
ISSN: | 1367-0751 |
Versão da editora: | https://academic.oup.com/jigpal/article/30/4/611/6123345 |
Arbitragem científica: | yes |
Acesso: | Acesso restrito UMinho |
Aparece nas coleções: | CAlg - Artigos em revistas internacionais / Papers in international journals |
Ficheiros deste registo:
Ficheiro | Descrição | Tamanho | Formato | |
---|---|---|---|---|
2102.12354.pdf Acesso restrito! | 2,55 MB | Adobe PDF | Ver/Abrir |